JP6738423B2 - Optical metrology in lithographic processes using asymmetric sub-resolution features to enhance measurements - Google Patents

Description

The present invention relates generally to lithography, and more specifically to metrology for lithographic processes.

A lithographic apparatus, such as a photolithographic apparatus, is a machine that applies a pattern onto a substrate, often a target portion on the substrate. A lithographic apparatus can be used, for example, in the manufacture of integrated circuits (ICs). In this case, patterning devices, which are alternatively referred to as masks or reticles, can be used to generate the circuit pattern to be formed on the individual layers of the IC. This pattern can be transferred onto a target portion (eg a field containing one or more die portions) on a substrate (eg a silicon wafer). The transfer of the pattern is typically via imaging onto a layer of radiation sensitive material (resist) provided on the substrate. In general, a single substrate will contain a set of adjacent target portions that are successively patterned.

In a lithographic process, it is desirable to frequently measure the resulting structures, for example for process control and monitoring. Various tools may perform such measurements, including, for example, optical tools such as various forms of scatterometers. These devices typically direct a beam of radiation onto a target (eg, a test structure on a patterned substrate) and measure the properties of the scattered radiation. Examples of such properties include the intensity of scattered radiation as a function of wavelength at a single reflection angle, the intensity as a function of reflection angle at one or more wavelengths, or the polarization as a function of reflection angle. The measured property often characterizes the diffraction "spectrum" from which the property of interest of the target can be determined.

The following is a non-exhaustive list of some aspects of the present technology. These and other aspects are described in the disclosure below.

Some aspects include a process of calibrating the model. This process obtains training data that includes scattered radiation information from multiple structures, each portion of the scattered radiation information being associated with a separate process condition that is characteristic of the patterning process of each structure. Calibrating the model with the training data by using one or more processors to determine a first ratio that relates a change in one of the process characteristics to a corresponding change in the scattered radiation information; including.

Some aspects include a process of estimating parameters of the patterning process. This process involves obtaining scattered radiation measurements of patterned structures on a substrate; using one or more processors to infer process characteristics of photolithographic patterning based on optical measurements using calibrated models. The model comprises a first ratio relating changes in one of the process characteristics to changes in the scattered radiation measurement.

The above and other aspects of the present technology will be better understood when the specification is read with reference to the following drawings, wherein like reference numerals indicate similar or identical parts.

FIG. 6 illustrates an example of a lithographic apparatus.

FIG. 6 illustrates an example of a lithographic cell or cluster in which the inspection technique according to the present invention may be used.

It is a figure which shows typically the operating principle of the spectral scatterometer as a 1st example of an inspection apparatus.

It is a figure which shows typically the angle-resolved scatterometer as another example of an inspection apparatus in schematic form.

5A and 5B are schematic diagrams of an inspection apparatus suitable for performing angle-resolved scatterometry and dark field imaging inspection.

FIG. 6 is a diagram schematically showing a target forming element on a reticle suitable for forming a grating having a focus-dependent asymmetry on a substrate.

FIG. 6 is a plan view showing an example of a test structure on a reticle according to some embodiments.

FIG. 8 is a cross-sectional view showing an example of a pattern test structure on a substrate corresponding to the reticle of FIG. 7.

8 is a flow chart showing an example of a process for calibrating a model based on the test structure of FIG. 7.

10 is a flow chart showing an example of a process for monitoring or controlling a lithographic process based on the model of FIG. 9.

FIG. 3 is a block diagram illustrating an example computer system in which certain steps of the techniques described above may be implemented.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will be described in detail herein. Drawings may not be to scale. However, the drawings and their detailed description are not intended to limit the invention to the particular forms disclosed, but rather to be within the spirit and scope of the invention as defined by the appended claims. It is to be understood that it is intended to cover variations, equivalents and alternatives of.

To reduce the problems described herein, we invented solutions, and in some cases, importantly, problems that were overlooked (or unforeseen) by others in the field of lithography. Had to recognize. In fact, we would like to emphasize the difficulty of recognizing these new issues that will become apparent in the future trend of the industry as we expect. Moreover, because multiple problems are addressed, some embodiments are problem-specific, and not all embodiments address every problem in conventional systems described herein, and , Does not provide all the benefits described in this document. Therefore, improvements that solve various variations of these problems are described below.

Before describing the embodiments of the present invention in detail, it is helpful to present an exemplary environment in which the present invention may be implemented.

FIG. 1 is a diagram schematically showing a lithographic apparatus LA. The apparatus is constructed to support an illumination system (illuminator) IL configured to condition a beam of radiation B (eg UV or DUV radiation); a patterning device (eg mask) MA, and patterning device according to specific parameters. A patterning device support or support structure (eg mask table) MT connected to a first positioning device PM configured to accurately position the substrate; each configured to hold a substrate (eg resist coated wafer) W. , Two substrate tables (eg wafer tables) WTa and WTb each connected to a second positioning device PW configured to accurately position the substrate according to certain parameters; A projection system (eg, a refractive projection lens system) PS configured to project the patterned pattern onto a target portion C (eg, including one or more dies) of a substrate W. The reference frame RF serves as a reference reference for connecting the various components and for setting and measuring the position of the patterning device and the substrate and the features thereon.

The illumination system may be refractive, reflective, magnetic, electromagnetic, electrostatic or other form of optical element, or any combination thereof, for directing, shaping, or controlling radiation. Various types of optical elements may be included.

In some embodiments, the patterning device support positions the patterning device in a manner that depends on the orientation of the patterning device, the design of the lithographic apparatus, and other conditions, such as whether the patterning device is held in a vacuum environment. Hold. The patterning device support can take many forms. The patterning device support may position the patterning device with respect to the projection system, for example.

The term "patterning device" as used herein should be broadly construed to refer to any device that can be used to pattern a cross section of a radiation beam to create a pattern in a target portion of a substrate. .. It should be noted that the pattern imparted to the radiation beam may not exactly match the desired pattern of the target portion of the substrate, eg if the pattern comprises phase shift features or so-called assist features. Generally, the pattern imparted to the radiation beam will correspond to a particular functional layer in the device being created in the target portion, such as an integrated circuit.

As shown, the device is transmissive (eg, using a transmissive patterning device). Alternatively, the device may be reflective (eg, using a programmable mirror array as described above or using a reflective mask). Examples of patterning devices include masks, programmable mirror arrays, programmable LCD panels. Any use of the terms "reticle" or "mask" herein may be considered synonymous with the more general term "patterning device." The term "patterning device" can also be interpreted as referring to a device that stores pattern information in digital form for use in controlling, such as a programmable patterning device.

As used herein, the term "projection system" refers to refraction, reflection, refraction, magnetic, electromagnetic, where appropriate for the exposure radiation used and other factors such as the use of immersion liquid or the use of vacuum. It should be broadly construed as encompassing any type of projection system, including mold and electrostatic optical systems or any combination thereof. Any use of the term "projection lens" herein may be considered as synonymous with the more general term "projection system."

The lithographic apparatus may be of a type in which at least part of the substrate is covered with a liquid having a relatively high refractive index (eg water) so that the space between the projection system and the substrate is filled. The immersion liquid may be applied to other spaces of the lithographic apparatus, for example between the mask and the projection system. Immersion techniques are expected to increase the numerical aperture of projection systems.

In operation, the illumination system IL receives a beam of radiation from a radiation source SO. The radiation source and the lithographic apparatus may be separate entities, for example when the radiation source is an excimer laser. In such a case, the radiation source is not considered to form part of the lithographic apparatus, and the radiation beam is directed from the radiation source SO towards the illuminator IL, for example a beam delivery system including suitable directing mirrors and/or beam expanders. Pass with the help of BD. In other cases, the source may be an integral part of the lithographic apparatus, for example when the source is a mercury lamp. The radiation source SO and the illuminator IL may optionally be referred to as a radiation system together with the beam delivery system BD.

The illuminator IL may include, for example, an adjuster AD for adjusting the angular intensity distribution of the radiation beam, an integrator IN, and a capacitor CO. An illuminator may be used to condition the beam of radiation and to have the desired uniformity and intensity distribution in its cross section.

The radiation beam B is incident on the patterning device MA, which is held on the patterning device support MT, and is patterned by the patterning device. After passing through the patterning device (eg mask) MA, the radiation beam B passes through a projection system PS which focuses the beam on a target portion C of the substrate W. With the help of a second positioning device PW and a position sensor IF (eg an interferometer device, a linear encoder or a capacitance sensor), for example the substrate table WTa or the substrate table WTa so that a different target portion C is located in the path of the radiation beam B. WTb can be moved accurately. Similarly, the first positioner PM and another position sensor (which is not explicitly shown in FIG. 1) are used to pattern the path of the radiation beam B, for example after a machine search from a mask library or during a scan. The device (eg reticle/mask) MA can be accurately positioned.

Patterning device (eg reticle/mask) MA and substrate W may be aligned using mask alignment marks M1, M2 and substrate alignment marks P1, P2. Although the substrate alignment marks are depicted as occupying dedicated target portions, they may be located in the spaces between the target portions (this is known as the scribe line alignment marks). Similarly, if the patterning device (eg mask) MA is provided with more than one die, the mask alignment marks may be located between the dies. Relatively small alignment marks may be included between device features inside the die, where the markers are as small as possible and require any imaging or process conditions that differ from adjacent features. It is desirable not to. The alignment system that detects the alignment marker is described separately below.

The depicted apparatus can be used in various modes. In scan mode, the pattern imparted to the radiation beam is projected onto the target portion C (ie a single dynamic exposure) while the patterning device support (eg mask table) MT and the substrate table WT are scanned synchronously. The speed and orientation of the substrate table WT relative to the patterning device support (eg mask table) MT can be determined by the magnification (reduction) and image reversal characteristics of the projection system PS. In scan mode, the maximum size of the exposure field limits the width (in the non-scan direction) of the target portion in a single dynamic exposure, while the length of the scan operation determines the height (in the scan direction) of the target portion. decide. Other types of lithographic apparatus and modes of operation are also possible, such as step mode. In so-called "maskless" lithography, the programmable patterning device is held stationary, but with a pattern change, the substrate table WT is moved or scanned.

The above described modes of use may be combined and/or modifications may be made to the above described modes of use or entirely different modes of use may be used.

The exemplary lithographic apparatus LA shown is of the so-called dual stage type with two substrate tables WTa, WTb and two stations, with which the substrate table can be exchanged between an exposure station EXP and a measurement station MEA. While one substrate on one substrate table is being exposed at the exposure station, another substrate can be loaded on the other substrate table at the measuring station and various preparatory steps are performed. This allows a substantial increase in device throughput. The preparing step may include mapping the contours of the surface of the substrate with the level sensor LS and measuring the position of the alignment marker on the substrate with the alignment sensor AS. If the position sensor IF cannot measure the position of the substrate table while the substrate table is in the measurement station or the exposure station, it is possible to track the position of the substrate table in both stations with respect to the reference frame RF. A position sensor may be provided. Other configurations may be used instead of the dual stage configuration shown. For example, another lithographic apparatus in which a substrate table and a measurement table are provided. They are docked together when preliminary measurements are performed, while they are detached when the substrate table is exposed.

As shown in FIG. 2, the lithographic apparatus LA, often also referred to as a lithocell or cluster, forms part of a lithographic cell LC that also includes apparatus for performing pre-exposure and post-exposure processes on a substrate. In most cases, these include a spin coater SC for depositing a resist layer, a developing device DE for developing the exposed resist, a cooling plate CH and a bake plate BK. The substrate handler or robot RO takes the substrate out of the input/output ports I/O1, I/O2, moves the substrate between different process devices and carries it to the loading bay LB of the lithographic apparatus. These devices, which are often collectively referred to as trucks, are under the control of the truck control unit TCU. The TCU itself is controlled by the supervisory control system SCS, which also controls the lithographic apparatus via the lithographic control unit LACU. Therefore, different devices can operate to maximize throughput and process efficiency.

Inspecting the exposed substrate by measuring characteristics such as overlay error with subsequent layers, line width, critical dimension (CD), etc. so that the substrate exposed by the lithographic apparatus is accurately and consistently exposed Is often desirable. Therefore, the manufacturing facility in which the lithocell LC is arranged also includes the metrology system MET that receives a part or all of the substrate W processed by the lithocell. The measurement result is directly or indirectly provided to the supervisory control system SCS. If an error is detected, adjustments may be made to the exposure of subsequent substrates, especially if the inspection can be performed quickly and quickly such that other substrates in the same batch are still exposed. Also, an already exposed substrate may be stripped and reworked for yield enhancement, or may be discarded, which allows further processing to be performed on substrates known to be defective. May be avoided. If only some target portions of the substrate are defective, then a further exposure can be performed only on the good target portions.

In the metrology system MET, the inspection device is used to determine the properties of the substrates, in particular how the properties of different substrates or different layers of the same substrate are different from layer to layer. The inspection apparatus may be integrated in the lithographic apparatus LA or the lithocell LC or may be a stand-alone apparatus. In order to speed up the measurement, it is desirable that the inspection apparatus measures the characteristics in the exposed resist immediately after the exposure. However, the latent image in the resist has a very low contrast, and there is no very small index difference between the resist areas that are exposed to radiation and the resist areas that are not exposed to radiation, so that all inspection systems have a latent image. It is not sensitive enough to perform an effective measurement of. Therefore, the measurement may be performed after the post-exposure bake step (PEB). PEB is usually the first step performed on the exposed substrate and increases the contrast between the exposed and unexposed parts of the resist. At this stage, the image in the resist may be referred to as the semi-latent image. It is also possible to carry out a measurement of the developed image (at which point either the exposed or the unexposed part of the resist has been removed) or after a pattern transfer step such as etching. It is also possible to execute. The latter possibility limits the possibility of reworking substrate defects, but may still provide valuable information.

FIG. 3 shows an example of a spectroscopic scatterometer that can be used as an inspection device in a measurement system of the type described above. In some embodiments, the scatterometer includes a broadband (white light) radiation projector 2 that projects radiation onto a substrate W. The reflected radiation goes to a spectroscope 4 which measures the spectrum 6 (intensity as a function of wavelength) of the specularly reflected radiation. From this data, the structure or profile 8 that gives rise to the detected spectrum is reconstructed by calculation in the processing unit PU (which may include one or more computer systems described below). This reconstruction can be performed, for example, by Rigorous Coupled Wave Analysis and non-linear regression methods, or by comparison with a library of pre-measured spectra or pre-computed simulated spectra. In most cases, because it is a reconstruction, the general form of the structure is known, and some parameters are of the process by which the structure was formed, except for the few parameters that should be determined from scatterometry data. Inferred from findings. Such a scatterometer may be configured as a normal incidence scatterometer or an oblique incidence scatterometer.

FIG. 4 illustrates example components of an angle resolved scatterometer that may be used in place of or in addition to the spectroscopic scatterometer. In this type of inspection device, the radiation emitted by the radiation source 11 is conditioned by the illumination system 12. For example, the illumination system 12 may include collimating with a lens system 12a, a color filter 12b, a polarizer 12c and an aperture device 13. The adjusted radiation travels along the illumination path IP, is reflected by the partially reflecting surface 15, and is focused on the spot S on the substrate W via the microscope objective lens 16. The measurement target T can be formed on the substrate. The lens 16 has a high numerical aperture (NA), preferably at least 0.9, more preferably at least 0.95. Immersion liquids can be used to obtain numerical apertures greater than 1 if desired.

As in the lithographic apparatus LA, one or more substrate tables may be provided to hold the substrate W during the measuring operation. The substrate table may be similar or identical to the configuration of the substrate tables WTa, WTb of FIG. 1 (in the example where the inspection apparatus is integrated in the lithographic apparatus, they may be the same substrate table). .. The coarse and fine positioner may be configured to accurately position the substrate with respect to the measurement optical system. Various sensors and actuators may be provided, for example, to obtain the position of the target of interest and to position the target of interest under the objective lens 16. Typically multiple measurements will be performed on the target at different locations across the substrate W. The substrate support can be moved in the X and Y directions to obtain different targets and in the Z direction to obtain the desired focus of the optical system on the targets. In practice, it is convenient to describe the operation so that the objective lens and the optical system are provided at different positions on the substrate even when only the substrate moves while the optical system remains substantially stationary. .. As long as the relative positions of the substrate and the optical system are accurate, it does not matter in principle whether one or both of them move in the real world.

When the radiation beam impinges on the beam splitter 16, in this example, part of it passes through the beam splitter and follows the reference path RP towards the reference mirror.

In this example, the radiation reflected by the substrate, including the radiation diffracted by any metrology target T, is collected by the lens 16 and travels the collection path CP towards the detector 19 through the partially reflecting surface. The detector may be arranged in the back projection pupil plane P located at the focal length F of the lens 16. In practice, the pupil plane itself may not be accessible and, instead, auxiliary optics (not shown) may be used to reimage it onto a detector located in the so-called conjugate pupil plane P′. The detector may be a two-dimensional detector capable of measuring the two-dimensional angular scattering spectrum or diffraction spectrum of the substrate target 30. In the pupil plane or the conjugate pupil plane, the radial position of the radiation defines the angle of incidence/exit of the radiation in the plane of the focus spot S, and the angular position around the optical axis O defines the azimuth angle of the radiation. The detector 19 may be, for example, an array of CCD or CMOS sensors and may use, for example, an integration time of 40 milliseconds per frame.

The radiation in the reference path RP may be projected on different parts of the same detector 19, or alternatively on different detectors (not shown). The reference beam is often used, for example, to measure the intensity of the incident radiation and to allow normalization of the intensity values measured in the scatter spectrum.

Various components of the lighting system 12 can be adjustable to achieve different metrology “recipe” within the same device. The color filter 12b may be implemented by a set of interference filters that select different wavelengths of interest within the range of, for example, 405-790 nm or shorter 200-300 nm. The interference filter may be modulatable rather than comprising a different set of filters. A grating can be used instead of the interference filter. The polarizer 12c may be rotatable or exchangeable to achieve different polarization states of the radiation spot S. The aperture device 13 can be adjusted to achieve different lighting profiles, as described below. The aperture device 13 is arranged in the plane P″ which is conjugate to the pupil plane P of the objective lens 16 and in the plane of the detector 19. In this way the illumination profile defined by the aperture device is different on the aperture device 13. It defines the angular distribution of radiation that passes through a location and is incident on the substrate.

The detector 19 may measure the intensity of the scattered light at a single wavelength (or narrow wavelength range), the intensity at multiple wavelengths separately, or the intensity integrated over a range of wavelengths. May be measured. Further, the detector may measure the TM and TE polarization intensities separately and/or may measure the TM and TE polarization phase differences.

If the metrology target T is provided on the substrate W, this may be a 1D grating printed after development such that the bars are formed of hard resist lines. The target may be a test structure designed to be easily measured by the scatterometer shown. The target may be a 2D grating in which the grating is printed after development such that the grating is formed with solid resist pillars or vias in the resist. The bars, pillars or vias may alternatively be etched into the substrate. This pattern is sensitive to the chromatic aberrations of the lithographic projection apparatus, in particular the projection system PS. The symmetry of the illumination and the presence of such aberrations will increase the variation itself in the printed grating. Therefore, the scatterometry data of the printed grating is used to reconstruct the shape of the grating. The parameters of the 1D grating, such as the line width and shape, and the parameters of the 2D grating, such as the width, length or shape of the pillars or vias, are implemented by the processing unit PU from the knowledge of the printing step and/or other scatterometry processes. It may be entered into the reconstruction process.

In addition to measuring parameters by reconstruction, angle-resolved scatterometry is useful for measuring feature asymmetry in the product and/or resist pattern. A specific application of asymmetry measurement is for the measurement of focus parameters (eg focus during exposure of the target) from a target printed with focus-dependent asymmetry. The concept of asymmetry measurement using the instrument of Figure 3 or 4 is described in US Patent Publication 20060066855, the contents of which are incorporated by reference. In most cases, the position of the diffraction order within the diffraction spectrum of the target is determined solely by the periodicity of the target, while the asymmetry of the intensity level within the diffraction spectrum accounts for the asymmetry within the individual features that make up the target. Show. In the instrument of FIG. 4, if the detector 19 can be an image sensor, such an asymmetry in the diffraction order will manifest itself directly as an asymmetry in the pupil image recorded by the detector. This asymmetry can be measured by digital image processing within the unit PU, from which focus can be determined in some embodiments.

FIG. 5A shows in more detail an inspection apparatus that has the same principle as the apparatus of FIG. 4, but realizes angle-resolved scatterometry with additional modifications for performing so-called dark field imaging. This device may be a stand-alone device, or may be incorporated into either the lithographic apparatus LA or the lithographic cell LC, eg at a measurement station. The optical axis with multiple branches through the device is indicated by the dotted line O. The target grating T and the diffracted rays are shown in more detail in Figure 5B.

The same reference numerals are used for the components already described in the device of FIG. The illumination path is labeled with IP as before. The reference path RP is omitted for clarity. Comparing the devices, the second beam splitter 17 splits the collection path into two branches. In the first measurement branch, the detector 19 records the scattering or diffraction spectrum of the target, exactly as described above. This detector 19 may be referred to as a pupil image detector.

In the second measurement branch, the imaging optics 22 forms an image of the target on the substrate W on the detector 23 (eg CCD or CMOS sensor). The aperture diaphragm 21 is provided in a plane conjugate with the pupil plane in the collection path (also referred to as a pupil diaphragm). The aperture stop 21 can take different forms, just as the illumination aperture can take different forms. Aperture stop 21 typically blocks the 0th order diffracted beam and functions so that the image of the target formed on sensor 23 is formed from only the 1st order beam. This is a so-called dark field image, which is similar to a dark field microscope. The images captured by the sensors 19 and 23 are output to the image processing controller PU. The function of the controller will depend on the particular type of measurement being performed.

In the illumination path of this example, additional optics are shown so that the field diaphragm 13 ′ can be placed in a plane that is conjugate to the plane of the target and image sensor 23. This plane may be referred to as the field plane or conjugate image plane and has a location where each spatial location across the field plane corresponds to a location across the target. This field stop is used, for example, to shape the illumination spot for a particular purpose, or to simply avoid illuminating features within the field of view of the device except for a portion of the target of interest. Good. Although the following figures and discussion refer to techniques for implementing the functionality of aperture device 13 for purposes of illustration, the present disclosure also includes the use of similar techniques to implement field aperture 13'.

As shown in more detail in FIG. 5B, the target grating T is arranged on the substrate W perpendicular to the optical axis O of the objective lens 16. In the case of an off-axis illumination profile, the illumination ray I is incident on the grating T from an angle away from the axis O and is a zero order ray (solid line 0) and two primary rays (dotted line +1 and dashed line -1). Cause It should be noted that when using a small overfilled target grating, these rays are only one of many parallel rays that cover the area of the substrate that contains the metrology target grating T and other features. is there. Since the aperture of the plate 13 has a finite width (sufficient to obtain a useful amount of light), the incident ray I actually occupies a certain angular range and the diffracted rays 0 and +1/-1 are to some extent. It will have spread. According to the point spread function of the small target, each order of +1 and -1 will spread further over a range of angles rather than a single ideal ray as shown.

Different illumination modes are possible by using different apertures. Each of apertures 13N (north) and 13S (south) provides off-axis illumination only from a particular narrow angular range. Returning to FIG. 5A, this is schematically illustrated by designating the diametrically opposite portions of the annular aperture north (N) and south (S). The +1st order diffracted ray (sign of +1(13N)) from the north part of the illumination cone enters the objective lens 16 and the −1st order diffraction order (sign of −1(13S)) from the south part of the cone. Is also the same. Using dark field image sensor 23 while switching apertures 13N, 13S of this type, as described in the earlier application mentioned in the introductory part, is one way to obtain asymmetry measurements from multiple small targets. Is the way. The aperture stop 21a can be used to block 0th order radiation when using off-axis illumination.

While off-axis illumination is shown, on-axis illumination of the target may be used instead and an aperture stop with an off-axis aperture is used to pass substantially only one of the first order diffracted light to the sensor. May be. In one example, a prism 21b having the effect of diverting the +1st order and the −1st order to different locations on the sensor 23 is used instead of the aperture stop 21 so that they can detect and compare without generating two images. it can. This technique is disclosed in the aforementioned US Patent Publication 20110102753, the contents of which are incorporated herein by reference. The secondary, tertiary and higher order beams (not shown in FIG. 5A or B) can be used for the measurement instead of or in addition to the primary beam.

When monitoring a lithographic process, it is often desirable to monitor the focus of the lithographic beam on the substrate. One known method of determining the focus setting from a printed structure is by measuring the critical dimension (CD) of the printed structure. CD is the minimum feature size (eg, element linewidth). The printed structure may be a target such as a line-space grating that is specially formed for focus monitoring. CD normally exhibits a quadratic response to focus, forming what is known as the "Bossung curve" on a plot of CD (y-axis) against focus (x-axis). The Bossung curve is a substantially symmetric curve, which is substantially symmetric about the peak showing optimum focus. Although this approach has some drawbacks, it does not imply that this approach cannot be used in some embodiments. One of the drawbacks is that it exhibits low sensitivity near optimum focus (due to the parabolic shape of the curve). Another drawback is that it does not react to any sign of defocus (because the curve is nearly symmetrical around optimal focus). This approach is particularly sensitive to dose and process variations (crosstalk).

In order to deal with these problems, a diffraction based focus (DBF) has been devised. Diffraction-based focusing can use targets that form features on a reticle that will print the target with some asymmetry depending on the focus setting during printing. This degree of asymmetry is a scatterometry method based on an inspection method by measuring the intensity asymmetry between the +1st and -1st order radiation diffracted from the target to obtain a measurement of the focus setting. Can be measured using.

FIG. 6 schematically illustrates a DBF test structure or target 615 configured for diffraction-based focus measurements. It comprises a plurality of DBF structures 620, each comprising a high resolution substructure 625. The high resolution substructure above the base pitch produces an asymmetric resist shape for each DBF structure 620, the degree of asymmetry being focus dependent. As a result, metrology tools can measure the degree of asymmetry from a target formed using the DBF target formation design 615, which can be translated into scanner focus.

While the DBF test structure 615 allows for diffraction-based focus measurements, it is not suitable for use in some situations. The thickness of the EUV resist film is significantly smaller than that used in immersion lithography, making it difficult to extract accurate asymmetry information from the asymmetry profile of the structures forming the part of the target. Also, such structures may not meet the strict design constraints applicable to a particular product structure. During the chip manufacturing process, all features on the reticle must be printed and survive subsequent processing steps. Semiconductor manufacturers use design rules as a means of constraining feature design to ensure that printed features meet their process requirements. Examples of such design rules relate to acceptable structure size or pitch. Another example design rule relates to pattern density, which may constrain the density of resist patterns to be located within a particular range. In some cases, these problems may be exacerbated by the use of the more widely adopted negative resist.

Some embodiments provide a relatively strong response to changes in focus, exposure, or other properties of the photolithography process, while providing test structures that meet commonly applied design rules. You can. Some embodiments provide sub-resolution features for patterning separation lines at the same focal point (relatively large depth of focus tolerance) or at relatively large exposure tolerances. Can be placed. In some cases, this test structure may facilitate the use of optical metrology, making it slower, more expensive, such as destructive and electron beam measurements, less information-intensive in certain ways, or even on production wafers. It may be preferable over other types of measurements that would otherwise be damaging (although it does not imply that these techniques are not used in some embodiments either). In some cases, the test structure may be mounted on a reticle and the measurement model may be: 1) patterning the substrate with the reticle while changing the conditions of the focus exposure matrix, 2) testing with a scatterometry tool. It may be calibrated to the test structure by optically measuring the structure and 3) fitting the model to the corresponding process conditions and optical measurements. This model may then be used to infer unknown process conditions experienced by the production substrate through optical measurements of the production substrate.

As described in more detail in some embodiments, the test structure includes a grating, with sub-resolution assist features extending parallel (eg, substantially parallel) to each bar of the grating asymmetrically away from the bar of the grating. Adjacent to each other. This sub-resolution feature is expected to have different effects on the sidewalls of the bar when patterned on the substrate depending on process conditions such as focus or exposure. The difference in sidewall shape is expected to provide a relatively robust signal to the optical metrology tool that is indicative of process conditions.

FIG. 7 schematically illustrates a portion of test structure 30 according to these techniques. In some cases, the test structure 30 resides within the measurement site grating on the reticle. In some embodiments, the test structure 30 includes a main feature 32, two sub-resolution assist features 34 and 36, and a sub-resolution inversion feature 38. Although the illustrated portion of test structure 30 is shown with dashed lines, this structure may be relatively uniform along its length in some embodiments. In some cases, portions of test structure 30 may be spaced and repeated in a parallel array to form a grating and the spacing of main features 32 may match the pitch of the grating. Sub-resolution assist features 34 and 36 may be spaced apart from main feature 32 by gaps 40 and 42, respectively.

The dimensions of test structure 30 may be selected to elicit various responses in patterned test structures on the substrate. Specifically, some dimensions may be selected that are applicable to optical measurements and elicit a response corresponding to photolithographic process conditions. The dimensions may be selected based on the wavelength of the radiation in patterning the test structure on the substrate or the wavelength of light in measuring the patterned test structure.

As shown in FIG. 7, the main feature 32 has a width 44 and the sub-resolution assist features 34 and 36 have widths 46 and 48, respectively. These widths may be smaller than the resolution limit of the photolithographic apparatus, which makes the feature "sub-resolution", although other dimensions of the feature may be larger than this limit. As a result, features 34 and 36 may not be transferred directly to the substrate during the lithographic patterning process.

However, features 34 and 36 may be dimensioned and spaced to affect the patterning of main features 32. In some cases, features 34 and 36 may be characterized as “scattering bars” or “sub-resolution assist features”. Sub-resolution features 34 may be spaced a distance 50 from main features 32, and sub-resolution features 36 may be spaced a distance 52 from main features 32. These distances may be selected such that features 34 and 36 affect the patterning of main features 32. To enhance this effect, in some embodiments the main feature may be a feature isolated to other instances of the test structure, eg, in the grating. In some cases, the distances 50 and 52 may be less than the wavelength of the radiation when lithographically patterning the test structure, eg, on the order of 10-150 nm in a 193 nm wavelength process, and proportional to other wavelengths. It may be smaller or larger so that In some embodiments, the pitch of the grating may be greater than this wavelength, for example greater than twice this wavelength.

Distances 50 and 52 can be different from each other, making features 34 and 36 asymmetric with respect to main feature 32. As a result, in some embodiments, sub-resolution features 34 and 36 may affect the transfer of main feature 32 when lithographically forming a pattern test structure. Due to the different distances 50 and 52, the effect could be different on different sides of the main feature. This will be described in more detail below with reference to FIG.

Sub-resolution inversion feature 38 may be dimensioned to elicit a similar effect. In some cases, the width 54 may be less than the lithographic resolution limit, so that the features 38 are not transferred to the substrate. The features 38 are located within the main feature 32 and are located at different distances 56 and 58 from opposite sides of the main feature 32. As a result, features 38 are expected to affect the transfer of patterns on different sides of main feature 32 differently, and will affect the right side to a greater extent than the left side in the orientation shown.

The wavelength of the radiation used in photolithographically transferring the pattern can guide the selection of other dimensions. In some embodiments, the width 44 of the main feature 32 is greater than the resolution limit of the light used in photolithographically transferring the pattern from the reticle to the substrate. Thus, the main feature may generate a corresponding pattern test structure on the substrate.

In some embodiments, the dimensions 52 and 50 and 56 and 58 are selected based on the desired amount of impact of the corresponding sub-resolution features 34, 38 or 36 on the adjacent sidewalls of the May feature 32. Good. In some cases, these effects are a function of both these individual distances and the process characteristics of the photolithography process that become apparent at the surface of the substrate (which actually develops when process drift occurs). (As opposed to process settings, which may differ from the processes that are done). For example, in some cases, depending on the shape of the test structure, the magnitude of the effect can be a function of the distance and one or more of focus, exposure, chromatic aberration, alignment, spherical aberration, or a combination thereof. In some cases, the effect of distance is proportional to changes in these process conditions, or in some embodiments, the effect of distance is not proportional to changes in these process conditions. In the non-proportional example, the difference between the sidewalls of the patterned main features is expected to indicate the process conditions that the substrate experiences, for example, the difference increases as the focus changes in one direction and the focus changes in the other direction. Can decrease as

Other light sources, such as those used in metrology, can guide the selection of several dimensions. In some embodiments, the pitch, eg, the spacing between instances of the main structure 32 in the grating, may be selected based on the light used to make the optical measurement. Similarly, the width 32 of the main feature may be selected based on the measurement light source.

In some embodiments, the test structure 30 may be a one-dimensional test structure where the distance 60 is substantially greater than the distance 44 (eg, an order of magnitude or more), and the shape of the test structure is compared over the distance 60. It may be uniform. Alternatively, in another embodiment, the shape may change or the shape may be interrupted. It should be noted that features 34, 36 and 38 are sub-resolution features at distance 60 greater than the resolution limit of the photolithographic apparatus.

It is expected that the shape of the test structure 30 will typically be applicable under the design rules expected to apply especially to negative resist processes. In the illustrated example, the sub-resolution features 34, 36 and 38 are parallel to the main feature 32 (eg, substantially parallel for purposes of comparing design rules and impact on the main features). This property is expected to reduce process integration issues for other test structures such as those described above with reference to Figure 6 where some features are orthogonal to the main features. That is, some embodiments use features such as those shown in FIG. 6 having orthogonal structures on one side of the main feature and the sub-resolution assist feature 34 of FIG. 7 on the other side, for example. Good.

The present technology is suitable for use with both positive and negative resists. When using a positive resist, the resist is exposed to radiation at the location where the underlying material is to be removed, and when using a negative resist the resist is exposed to radiation at the location where the underlying material is to be stored, For example, subsequent etching follows. The terms "main feature" and "assist feature" apply to both areas. In some cases, these features may correspond to a portion of the reticle that blocks light, or depending on the type of resist used, or one of the reticles that allows radiation to be transmitted toward the substrate. It may correspond to a section.

FIG. 8 schematically shows a simulated resist profile produced on a substrate after lithographic patterning by the test structure of FIG. The illustrated pattern test structure 65 is shown in cross-section with dimension 60 normal to the plane of FIG. The illustrated pattern test structure 65 includes two bodies 66 and 68 of photoresist on either side of a trench 70 in the resist. These structures may be on a substrate that underlies region 71. The trench 70 may correspond to the main feature 32 of FIG. In particular, assist features 34, 38 and 36 are not shown directly or partially in the pattern test structure of FIG. This is because these features 34, 38 and 36 are smaller than the resolution limit of the photolithography process.

However, features 34, 38 and 36 affect the shape of trench 70 in pattern test structure 65. Trench 70 is defined in part by sidewalls 72 and 74. The sidewalls rise with respect to the normal vector of the substrate 71 at angles 76 and 78, respectively. As shown, angles 76 and 78 are different, with angle 78 being greater than angle 76 in this example. This difference is believed to be due in part to the effect of the sub-resolution features 34, 36 and 38 and the relative distance to the respective sidewall of the main feature 32 of FIG. In some embodiments, the difference between angles 76 and 78 can be affected by the process characteristics of the photolithography process when pattern test structure 65 is transferred to substrate 71.

Often, these process characteristics are specified within the settings of the photolithography apparatus, but the actual process characteristics experienced by the surface of the substrate 71 often deviate from this setting. For example, characteristics such as focus and exposure, although small in amount, can drift in a relatively difficult way to predict over time that may be relevant. Any number of factors may cause drift in the process characteristics, such as heating of the lens of the lithographic apparatus, changes in the environment of the manufacturing equipment, changes in the underlying stack, consumption of consumable components, etc.

In some embodiments, such drift can be detected optically by measuring the test structure 65 using an ellipsometer, such as a scatterometer. The test structure 65 can be repeated in a grid or grating above the measurement location on the substrate, and the scatterometry techniques described above can be used to measure the shape characteristics of the test structure 65. In some cases, these measurements include signals indicative of angles 76 and 78, and may include, for example, measurements indicative of the difference between angles 76 and 78. In some cases, such optical measurements can be relatively fast and inexpensive compared to other techniques that can characterize test structures. For example, top-down scanning electron microscopy measurements often do not provide sufficient information about the shape of sidewalls 72 and 74, and can be relatively slow and expensive. Similarly, obtaining a cross section of a substrate is a relatively slow process to produce, which can be problematic because it can destroy the substrate. However, this does not suggest that some embodiments may not be used in combination with these measurement techniques.

In some cases, the shape of pattern test structure 65 may depend on both optical and non-optical effects. For example, the lateral stress in resists 66 and 58 created by resist shrinkage may cause a shape shift of angles 72 and 74, as the resist attached to the substrate shrinks more than substrate 71, and , These effects may also be measured in some embodiments.

In some embodiments, the optical measurements of the pattern test structure 65 can be correlated to process conditions that tend to produce its shape, eg, in the correlation model or mechanical model described below. In some embodiments, after calibration, the model can be used to monitor the photolithography process based on observed optical measurements.

It should be noted that the test structure of FIG. 7 and the pattern test structure 65 of FIG. 8 are merely examples of the present technology. Various other configurations may be used. For example, additional sub-resolution features 34 and 36 may be placed at different distances on either side of the main feature 32 to enhance the effect of the sub-resolution features. In some embodiments, sub-resolution features 34 or 36 may include additional sub-resolution features on one side thereof, but not on the other side. In some cases, one sub-resolution feature may be symmetric while another is asymmetric. In some cases, three or more sub-resolution features may be located on one side and absent on the other. Similar variations can be applied to sub-resolution inversion feature 38. For example, multiple sub-resolution inversion features 38 may be located at various locations within the main feature 32. In some cases, these features may be positioned asymmetrically with respect to the main feature 32 and may affect the sidewalls of the pattern test structure 65 differently. The illustrated test structure 30 is straight and continuous (eg, substantially straight and continuous) along the dimension 60, although in other embodiments the pattern test structure 30 may include curved or discontinuous portions. In some cases, the main features may have other shapes, for example, the main features may be pillars or vias, and the assist features may extend around the perimeter of the pillars or vias. In that case, it may be asymmetric with respect to the central axis of the pillar or via. Various other variations are consistent with these examples.

FIG. 9 schematically illustrates an example of a process 80 for calibrating a model that infers lithographic process characteristics (e.g. revealed at the surface of the substrate) based on optical measurements of the test structure of the above example, for example. In some embodiments, the process 80 intentionally changes the process characteristics of photolithography through a parameter space (eg, through a matrix of values) and observes the patterned test structure under varying conditions. Correlating the process settings with the measured optical values. Such correlation may take various forms, including training a machine learning model, selecting model parameters, setting up a look-up table, and the like.

In some embodiments, the process 80 includes a main feature and one or more proximate to the main feature, asymmetric with respect to the main feature, and substantially parallel to the main feature, as indicated by block 82. Acquiring a reticle having a test structure with sub-resolution features. In some embodiments, the reticle may be the reticle having the test structure of FIG. 7 or one or more variations of the above. In some embodiments, obtaining the reticle comprises obtaining a photolithography apparatus in which the reticle is mounted, eg, a photolithography apparatus in a semiconductor manufacturing plant in which electronic devices are manufactured using the reticle. May be included. In some cases, the test structure may be placed on a reticle with a layout of layers for functional devices such as integrated circuits. For example, a semiconductor manufacturing plant may include multiple reticles in multiple photolithographic apparatus for patterning different layers of a semiconductor device, and different reticles may be used to pattern different layers.

Next, some embodiments may photolithographically pattern the substrate while changing the process characteristics of the photolithographic process, as indicated by block 84. Patterning the substrate may include patterning one or more substrates depending on the desired sample size. Photolithographic patterning may include patterning a pattern test structure as shown in FIG. 8 to form on the surface of the substrate using the photolithographic apparatus described above. Changing the process characteristics may include changing one or more photolithographic process characteristics over a range of values. For example, the amount of increase over that range may be fixed or may change. In some cases, multiple process characteristics may vary over multiple discrete ranges. For example, focus and exposure settings may vary over discrete ranges according to a focus-exposure dose matrix. In some cases, additional process characteristics may be varied in higher order matrices. In some cases, different exposure fields on the substrate may be treated differently according to different process characteristic changes. For example, a first field containing multiple dies, etc., may be patterned with a first set of focus and exposure settings, and then a second different field containing different instances of multiple dies, etc., with different focus. It may be patterned with different settings, different exposure settings, or both. In some cases, the test structures described above may be located within scribe lines between individual dies or within individual dies. In some embodiments, photolithographic patterning may include developing and baking the photoresist prior to measurement.

Next, some embodiments may optically measure the pattern test structure, as indicated by block 86. Optical measurements may be performed using one or more of the scatterometry techniques described above. In some cases, multiple pattern test structures within each field may be measured to obtain results for each processing condition with a more stable sample size. In some embodiments, the optical measurement comprises receiving an optical signal indicative of a shape of the first sidewall of the main feature and receiving an optical signal indicative of a shape of a different (second) sidewall of the main feature. Then determining the difference between the different optical signals to calculate a derivative of the optical measurement. In some cases, the optical measurements for each treatment condition may be aggregated using a representative value such as, for example, the mean, mode or median, or in some embodiments the values may be Instead of being aggregated, it may be used to enhance various machine learning techniques using, for example, cross validation or bootstrap aggregation.

Next, some embodiments may calibrate the model based on the test conditions and corresponding optical measurements, as indicated by block 88. In some cases, the calibration record is formed using a plurality of test conditions, each test condition including a setpoint for a photolithography process using the test condition, and the test conditions. And a set of optical measurements of the pattern test structure. A variety of different models and calibration techniques may be used. In some cases, the model may be a look-up table having an index corresponding to a range of optical measurements and a value corresponding to the test condition observed within that range to generate the optical measurements. .. In another example, the model may be created by training a decision tree with test data. In some cases, calibrating the model based on the test conditions may include, for example, updating the pre-existing model with Bayesian estimation to account for the newly observed test data. In some cases, calibrating the model may include calibrating the Hidden Markov Model with the process characteristics hidden. In some cases, the model may be calibrated multiple different times with different subsets of test data, and different instances may be compared to each other to improve the reliability of the model, Alternatively, they may be aggregated by cross validation.

この式（「式１」）において、「Ｐ」は、例えば特定の範囲の波長または角度にて、散乱計測の瞳で観測される光学信号、例えば光強度であり；ドーズ（dose）は、基板の表面での放射のドーズであり；フォーカス（focus）は、基板の表面でのフォーカス条件であり；「…」は、複数項のパターンにわたって拡張する高次の項を表し、これはより小さい影響に起因していくつかの場合に省略されてもよいが、他の特徴も省略されてもよいことを示唆するものではない。 In some embodiments, the model may be based on the following Taylor series expansion:

In this equation (“Equation 1”), “P” is the optical signal, eg, light intensity, observed at the pupil of the scatterometry, eg, at a wavelength or angle in a particular range; the dose is the substrate Is the dose of radiation at the surface of the substrate; focus is the focus condition at the surface of the substrate; "..." represents the higher order terms that extend over the multi-term pattern, which has a smaller effect. May be omitted in some cases due to, but does not imply that other features may also be omitted.

Using Equation 1 and scatterometry tool pupil measurements, some embodiments (e.g., error function, e.g., the root mean square of the conditions predicted by the model for the settings in the focus-exposure matrix). The parameters of the equation may be chosen (by choosing partial differential terms that tend to reduce the root sum). In some cases, the model may be trained with gradient descent, such as stochastic gradient descent. In some embodiments, the model may be trained on multiple parameters, for example, some test structures may be tested for exposure and the remaining test structures may be tested for focus, or the model may be tested. May be trained on both process characteristics. In some cases, the optical signal is an actuation signal obtained by the difference between two pupil plane measurements to calibrate the focus shift relative to the pupil pixel shift.

The model based on Equation 1 can take many forms. In some cases, the model is a sub-expression 1 itself with coefficients determined based on the training data set, such as the first three terms, the first four terms, the five terms, the six terms, or more of the equations. Is. In some cases, the model includes several versions of Equation 1, each solved for different process characteristics. In some cases, the model is a look-up table that approximates Equation 1, or a trained machine learning model that approximates Equation 1, such as a trained decision tree or neural net that fits the universal approximation theorem. And so on.

In some embodiments, the model is built with a Markov chain Monte Carlo algorithm, such as the Metropolis-Hasting algorithm, and executed by one or more computing devices described below. In some embodiments, such algorithms may, at run-time, iteratively build Markov chains based on samples that tend to be drawn from relatively high probability regions of the desired probability distribution. Some embodiments may include an initialization step in which points in the parameter space and probability densities are randomly selected, eg, for the points, resulting in a Gaussian distribution centered on the points selected for the probability densities. To be selected. Some embodiments then iterate to 1) determine the next candidate point in the parameter space based on the probability density by selecting subsequent points by sampling the probability density, and 2) Determine a pass rate that indicates whether a candidate point is more likely than the current point, and 3) accept the next candidate point in response to determining that the point is more likely, or the probability of that point. The next candidate point may either be rejected depending on the determination that is low. Passed points can form Markov chains in the parameter space and tend to randomly walk towards a relatively high probability region of that space and stay within that region. In some embodiments, the initial state threshold amount of the chain may be discarded during the "burn-in" period during which the chain moves to a higher probability region. The resulting Markov chain points (eg, high dimensional vectors) may be used to calculate expected values for various process characteristics of the lithographic process using Monte Carlo analysis.

The calibration determines for measurement within the training set the coefficients of Equation 1 that match the value P to the focus, dose and other process characteristic settings that are intentionally changed when patterning the test structure for the training data set. May be included. In some cases, the match may be sign-dependently characterized by an error function or a fitness function that aggregates the error over a set of test conditions on the training data set, such as the sum of root mean squares. Some embodiments may increase the match between the model and the training set until the change between successive iterations is less than a threshold, for example according to the partial derivative of each parameter with respect to the measure of the match. The parameters that tend to decrease may be selected repeatedly. Some embodiments may iterate this process from different initial selections of parameters and select the result of the iteration that produces the closest match between the training set and the model to prevent the selection of local minima.

In some embodiments, determining the coefficient of Equation 1 includes determining a ratio that relates a change in one of the process characteristics to a change in pupil intensity. This may include determining a ratio associated with pupil intensity for changes in only one process characteristic. In some embodiments, determining the ratio may also include determining a mixed derivative that relates changes in the subset of process characteristics to changes in pupil intensity. Some embodiments may also include determining a total derivative for pupil intensity. Determining the derivative (derivative) determines the incremental change in one process characteristic, for example, without changing the target value in the lithographic apparatus, while fixing the other process characteristic substantially constant, and the measured change in pupil intensity. May be approximated by relating to Determining the derivative may also include determining the reciprocal of the derivative.

In some embodiments, once the model is calibrated, the model may be uploaded to the process monitoring module of the lithographic apparatus. This module may include one or more of the computer-executed code described below, may perform the processes described below with reference to FIG. 10, and may perform process control or monitoring of the photolithography process.

In some embodiments, the process 89 includes a main feature and one or more proximate to the main feature, asymmetric with respect to the main feature, and substantially parallel to the main feature, as indicated by block 90. Acquiring a reticle having a test structure with sub-resolution features. In some cases this may be the same as the reticle described above. However, the processes of FIGS. 9 and 10 are not limited to the test structures described above with reference to FIGS. 7-8 (which does not imply that other features are similarly limited). In fact, these processes are also particularly suitable for use with the test structure of FIG.

Next, some embodiments photolithographically pattern a manufacturing substrate (eg, one substrate has a manufacturing facility widget as opposed to a test substrate), as indicated by block 91, using a reticle. You can do it. In some embodiments, this step may include, for example, patterning, or partially patterning, a plurality of production substrates within a set of production substrates.

Next, some embodiments may optically measure the pattern test structures on the production substrate, as indicated by block 92. In some cases, a subset of manufacturing substrates may be selected for measurements that are representative of the overall manufacturing process, or in some cases each substrate in the manufacturing process may be measured. In some cases, the optical measurement may be similar or identical to that described above with reference to FIG. In some embodiments, the optical measurement is a scatterometry measurement, and in some cases, the scatterometry measurement produces a difference value indicative of a difference in sidewall shape of the pattern test structure. In some embodiments, multiple locations within the substrate may be measured, eg, multiple locations within each exposure field or within each die.

Next, some embodiments may infer (eg, through indirect measurements) the process characteristics of photolithographic patterning based on optical measurements and models, as indicated by block 93. In some embodiments, this estimation may include inputting optical measurements into the model calibrated in the process of FIG. 9 and outputting estimated process characteristics. In some cases, the inferred process characteristic is an absolute value, such as focus, exposure, chromatic aberration, overlay, etc. Or, in some cases, the inferred process characteristic is a difference, such as a change in one or more of these values. Inferring process characteristics may include inferring a ratio of process characteristics or some other combination of process characteristics. In some cases, inferring process characteristics may include inferring multiple process characteristics using one model or multiple models. In some cases, the process characteristics may be inferred for each of a plurality of different locations on the substrate, eg, each field of the substrate, each die of the substrate, each region of each die of the substrate. ..

Next, some embodiments may determine whether the inferred process characteristic differs from a target value, as indicated by block 94. In some cases, this step may include determining whether the inferred process characteristic falls within a threshold of a range of acceptable values, or in some cases, this step may include determining that some May be simply determined at different resolutions. In some cases, for example, for a substrate field or die, it may include determining whether a measurement site that is greater than a threshold amount (eg, count or ratio) has a process characteristic that differs from a target value. In some cases, different thresholds may be applied to different areas of the substrate, field or die, for example, to prevent rework due to infrequently expected edge die measurements. In some cases, comparing a process characteristic to a target value may be accomplished by, for example, calculating a score based on the process characteristic corresponding to an acceptable target value and a cross product of the target values, thereby obtaining a plurality of target values. Multiple process characteristics may be compared against.

Next, some embodiments may adjust the settings of the photolithographic patterning process to reduce the difference from the target value, as indicated by block 95. In some cases, this step may include adjusting the focus or exposure of the photolithographic apparatus to reduce the difference. In some embodiments, this adjustment reduces drift in process conditions that, if not adjusted, can remain undetected for extended periods of time while additional substrates are being processed below allowed process conditions. Good. In some embodiments, this step may additionally or alternatively include reworking the substrate, such as by removing resist and repatterning.

In some embodiments, this process 89 may be repeated multiple times at various layers within a semiconductor manufacturing plant where electronic devices are manufactured. For example, a substrate undergoes multiple cycles of deposition/implantation, patterning, etching, etc. to build various layers of a semiconductor device, then singulated from the substrate, packaged, and placed in an electronic device. Can be done.

FIG. 11 schematically illustrates a computer system 100 that can assist in implementing the methods and flows disclosed herein. Computer system 100 includes a bus 102 or other communication means for communicating information, a processor 104 (or multiple processors 104 and 105) coupled to bus 102 for processing information. Computer system 100 also includes a main memory 106, such as a random access memory (RAM) or other dynamic storage device, connected to bus 102 for storing information and instructions to be executed by processor 104. Main memory 106 may be used to store temporary variables and other intermediate information during execution of instructions to be executed by processor 104. Computer system 100 further includes a read only memory (ROM) or other static storage device connected to bus 102 for storing static information and instructions for processor 104. A storage device 110, such as a magnetic disk or optical disk, is provided and is connected to the bus 102 for storing information and instructions.

Computer system 100 may be connected via bus 102 to a display 112 such as a cathode ray tube (CRT) or flat panel or touch panel display for displaying information to a computer user. Input devices 114, including alphanumeric and other keys, are connected to bus 102 for information communication and command selection with processor 104. Another type of user input device is a cursor controller 116, such as a mouse, trackball or cursor direction keys for communicating directional information and command selection to the processor 104 and controlling cursor movement on the display 112. The input device typically has two degrees of freedom, a two-axis, a first axis (eg x) and a second axis (eg y), allowing the device to specify a position on a plane. A touch panel (screen) display may be used as an input device.

According to one embodiment, some of the processes described herein are performed by computer system 100 in response to processor 104 executing one or more sequences of one or more instructions contained in main memory 106. May be done. Such instructions may be read into main memory 106 from another computer-readable medium, such as storage device 110. Execution of the sequences of instructions contained in main memory 106 causes processor 104 to perform the process steps described herein. One or more processors may be employed in the multi-processing configuration to execute the sequences of instructions contained in main memory 106. In alternative embodiments, hardware implemented circuitry may be used in place of, or in combination with, software instructions. Therefore, the description herein is not limited to any particular combination of hardware circuitry and software.

Various forms of computer readable media may be involved in carrying one or more sequences of one or more instructions to processor 104 for execution. For example, the instructions may first be generated on a computer-readable medium of a remote computer. The remote computer can load the command into its dynamic memory and send the command via wired or wireless communication. Computer system 100 can receive the instructions and place the instructions on bus 102. The bus 102 carries the command to the main memory 106, and the processor 104 retrieves the command from the main memory 106 and executes the command. The instructions received by main memory 106 may be selectively stored on storage device 110 either before or after execution by processor 104.

In one embodiment, computer system 100 includes a communication interface 118 connected to bus 102. Communication interface 118 provides two-way data communication that connects to a network link 120 that connects to a local network 122. For example, communication interface 118 may be an integrated services digital network (ISDN) card or modem that provides a data communication connection to a corresponding communication line. As another example, the communication interface 118 may be a LAN card that provides a data communication connection to a compatible local area network (LAN). Wireless links may be implemented. In any such implementation, communication interface 118 sends and receives electrical, electromagnetic or optical signals that carry digital data streams representing various types of information.

Network link 120 typically provides data communication through one or more networks to other data devices. For example, the network link 120 may provide a connection to a host computer 124 through a local network 122, or a data facility executed by an Internet service provider (ISP) 126. The ISP 126 in turn provides data communication services through the world wide packet data communication network now commonly referred to as the “Internet” 128. Both the local network 122 and the Internet 128 use electrical, electromagnetic or optical signals that carry digital data streams. Signals through various networks for transmitting and receiving digital data to computer system 100 and through communication interface 118 on network link 120 are exemplary forms of carrier waves for transmitting information.

Computer system 100 can send messages and receive data, including program code, through the networks, network links 120 and communication interfaces 118. In the Internet example, server 130 may send a request code for an application program through Internet 128, ISP 126, local network 122 and communication interface 118. One of the applications thus downloaded may provide, for example, performance of some of the processes of certain embodiments described herein. The received code may be executed by processor 104 as it is received, and/or stored in storage device 110 or other non-volatile storage for later execution. In this way, the computer 110 may obtain the application code in the form of carrier waves.

The disclosed embodiments may be implemented in hardware, firmware, software, or a combination thereof. Embodiments of the disclosure may be implemented as instructions stored on a machine-readable medium or a computer-readable medium, which may be read and executed by one or more processors. The term "machine-readable medium" or "computer-readable medium" as used herein refers to any medium that participates in providing instructions to a processor 104 for execution and/or a machine (eg, a computing device). ) Include any mechanism configured to store or transmit information in a readable form. Such a medium may take various forms, including but not limited to, non-volatile, non-transitory media, volatile, non-transitory media, and transmission media. Non-volatile media includes, for example, optical or magnetic disks, such as storage device 110. Volatile media includes dynamic memory, such as main memory 106. Transmission media includes coaxial cables, copper wire and fiber optics, including the wires that comprise bus 102. Transmission media can also take the form of acoustic or light waves, such as those generated during radio frequency (RF) and infrared (IR) data communications. Common forms of computer readable media are, for example, floppy disks, flexible disks, hard disks, magnetic tape, magnetic disk storage media or other magnetic media, CD-ROM, DVD or any other optical storage media, punched cards, paper tape or Physical media with other hole patterns, read only memory (ROM), random access memory (RAM), PROM, EPROM, flash EPROM, flash memory device or any other memory chip, or cartridge, carrier wave (eg, It includes an electrical, optical, acoustic or other form of transmission signal (eg, carrier wave, infrared signal, digital signal, etc.) or any other computer-readable medium.

In addition, firmware, software, routines, instructions may be described herein as performing certain actions. However, such a description is merely for convenience and such acts may actually occur as a result of a computing device, processor, controller or other device executing firmware, software, routines, instructions, etc. Be understood.

Although described herein as an example of the use of a lithographic apparatus in the manufacture of ICs, the lithographic apparatus described herein may include, for example, integrated optical systems, magnetic domain memory guiding and sensing patterns, flat panel displays, liquid crystal displays (LCDs). It will be appreciated that it may also have other applications, such as the manufacture of thin film magnetic heads. Those skilled in the art will recognize that in such alternative applications any use of the terms "wafer" or "die" herein is synonymous with the more general term "substrate" or "target portion", respectively. It will be understood that this can be considered. Substrates referred to herein may be pre- or post-exposure by, for example, a track (typically a device that applies a resist layer to the substrate and develops post-exposure resist), metrology tools, and/or inspection tools. It may be processed. Where applicable, the disclosure herein may be applied to these or other substrate processing apparatus. The substrate may also be processed multiple times, for example to produce a multi-layer IC, in which case the term substrate in this document may also mean a substrate containing a number of processing layers that have already been processed.

The terms "radiation" and "beam" as used herein include any type of electromagnetic radiation, including ultraviolet (UV) radiation (eg, having a wavelength of 365 nm, 248 nm, 193 nm, 157 nm or 126 nm), polar. Includes short ultraviolet (EUV) radiation (eg, having a wavelength in the range of 5-20 nm) and particle beams such as ion and electron beams.

Although the components illustrated in the block diagrams are depicted as separate functional blocks, embodiments are not limited to systems in which the functions described herein are organized as illustrated. .. The functionality provided by each component may be provided by software or hardware that is organized differently than what is currently illustrated, such as such software or hardware (e.g., in a data center or They may be (geographically) mixed, combined, duplicated, divided, distributed, or otherwise organized differently. The functionality described herein may be provided by one or more processors on one or more computers executing code stored on tangible, non-transitory, machine-readable media. In some cases, a third party content delivery network may host some or all of the information delivered through the network, in which case, to some extent, until information (eg, content) is said to have been provided or provided, The information may be provided by sending a command to retrieve the information from the content delivery network.

The reader will understand that this application discloses multiple inventions. Instead of dividing the inventions into separate patent applications, the applicants grouped these inventions into a single document because the relevant subject matter applies to the economics of the filing process. is there. However, the separate advantages and aspects of such an invention must be consistent. In some cases, embodiments address all of the shortcomings described herein, but the invention is independently beneficial, and some embodiments address only a subset of such problems. It will be appreciated that, or provide other unreferenced benefits that will be apparent to those of skill in the art upon reviewing this disclosure. Due to cost constraints, some inventions described herein may not be presently claimed, in later applications such as continuation applications, or by amending the present claims. May be stated in the claim. Similarly, due to space constraints, either the summary of this document or the summary of the invention section contains a comprehensive list of all such inventions or all aspects of such inventions. It should not be interpreted.

The description and drawings are not intended to limit the invention to the particular forms disclosed, but on the contrary, all changes and equivalents within the spirit and scope of the invention as defined by the appended claims. It is to be understood that it is intended to cover things and alternatives. Further modifications and alternative embodiments of various aspects of the invention will be apparent to those of skill in the art given this description. Therefore, this description and drawings are to be construed as illustrative only and are for the purpose of teaching those skilled in the art the general manner of carrying out the invention. It will be appreciated that the forms of the invention shown and described herein are to be construed as examples of embodiments. Subsequent to having the benefit of this description of this document, alternative elements and materials to those shown and described herein may be used, as would be understood by one of ordinary skill in the art, and portions thereof. And the process may be reversed or omitted, and particular features of the invention may be utilized independently. Changes may be made in the elements described herein without departing from the spirit and scope of the invention as described in the following claims. The headings used herein are for organizational purposes only and are not meant to be used to limit the scope of the description.

As used throughout this application, the words "may" and "may" have an admissible meaning (i.e., possible) rather than a compulsory meaning (i.e., must mean). (Meaning to have sex). The word "comprising" means including, but not limited to. As used throughout this application, the singular includes the plural unless the context clearly indicates otherwise. Thus, for example, reference to "an element" includes two or more elements, despite the use of other terms or phrases for one or more elements, such as "one or more." The term “or/or” is non-exclusive unless stated otherwise, ie it includes both “and/and” and “or/or”. Terms such as “Y depending on X”, “Y when X, “Y” if “X”, “Y in case of X, Y”, etc. are necessary conditions for the predecessor. It includes a causal relationship that becomes a sufficient condition or a contributory causal condition that the antecedent results in. For example, "state X occurs when condition Y is acquired" means "X occurs only when Y" and " X occurs when Y and Z. Such conditional relationships are not limited to results that immediately follow the preceding acquisition, as some results may be delayed, and conditional statements can be associated with the result by the antecedent, for example by the antecedent. Related to sex. Descriptions in which multiple features or functions are associated with multiple objects (eg, one or more processors that perform steps A, B, C, and D) refer to all such features or functions unless otherwise indicated. Both being associated with all such objects and having a subset of properties or functions associated with a subset of properties or functions (eg, all processors perform steps A-D respectively). Both when executing, when processor 1 executes step A, when processor 2 executes part of step B and step C, and when processor 3 executes part of step C and step D). To do. Further, unless stated otherwise, the description that one value or action is "based on" another condition or value means that the condition or value is the only factor and that the condition or value is multiple factors. Both cases where it is one factor are included. Unless stated otherwise, the statement that each "individual/each" of an instance of some set has some property means that some of the larger set, which are not the same or similar components, do not have that property. It should not be read as excluding, that is, each does not have to mean each and every. Unless otherwise stated, as will be apparent from the discussion, the use of terms such as "process," "compute," "compute," "determine," etc., consistent with the discussion in this specification refers to special-purpose computers or It will be appreciated that it refers to the operation or process of a particular device, such as a similar special purpose electronic processing/computing device.

In this patent, certain US patents, US patent applications or other sources (eg, articles) are incorporated by reference. However, the text of such US patents, US patent applications, and other sources is incorporated by reference to the extent there is no conflict between such sources and the description and drawings shown herein. When such conflicts arise, no part of such conflicting text in US patents, US patent applications, and other materials incorporated by such reference is expressly incorporated by reference in this patent.

The technology will be better understood with reference to the sections listed below.
(Claim 1) A method of calibrating a model, the method comprising: scattered radiation information from a plurality of structures, each portion of the scattered radiation information in a separate process condition that is characteristic of the patterning process of each structure. By obtaining training data containing relevant scattered radiation information; by using one or more processors to determine a first ratio that relates a change in one of the process characteristics to a corresponding change in the scattered radiation information. Calibrating the model with the training data.
(Item 2) The method of item 1, wherein the calibrating comprises determining a second ratio that relates a change in another one of the process characteristics to a change in scattered radiation information.
(Claim 3) The method according to Item 2, wherein the first ratio is a partial derivative or reciprocal of focus with respect to scattered radiation information, and the second ratio is a partial derivative or reciprocal of dose with respect to scattered radiation information. is there.
(Item 4) The method according to any one of items 1-3, wherein the model is based on at least three Taylor series expansions.
(Item 5) The method according to any one of items 1-4, wherein the model includes a second derivative of the process characteristic with respect to scattered radiation information or an inverse of the second derivative.
(Claim 6) In any one of paragraphs 1-5, calibrating the model comprises determining model parameters using a Markov chain Monte Carlo algorithm.
Clause 7: The method of any of clauses 1-6, wherein calibrating the model comprises determining model parameters using a Metropolis Hastings algorithm.
(Claim 8) The method of any of clauses 1-7, wherein calibrating the model comprises adjusting the model parameters iteratively between each iteration of the model and at least a portion of the training data. Determining the model parameter based on the aggregated value of the conformances.
(Item 9) The method according to any one of items 1-8, wherein the scattered radiation information is used to pattern a plurality of test structures on one or more substrates according to a focus-exposure dose matrix. Of the plurality of test conditions showing the process characteristics of the test pattern, in which the test structures having different patterns are patterned under different test conditions, and the pupil intensity measurement value by scatter measurement of the plurality of pattern test structures. Data indicating which test conditions the measurements correspond to, the model infers characteristics of the photolithography process from optical measurements of test structures patterned in the photolithography process.
(Item 10) The method according to any one of items 1-9, wherein the training data comprises simulated training data acquired based on a model of a photolithography process and a model of scattered radiation information.
(Item 11) The method according to any one of items 1-10, wherein the main feature and one or more sub-resolution features adjacent to the main feature on the reticle and substantially parallel to the main feature are included. Obtaining a reticle that at least partially defines a test structure; photolithographically patterning one or more substrates with the reticle.
Clause 12: The method of clause 11, wherein the one or more sub-resolution features are on the first side of the main feature a first distance dimensioned to affect a corresponding pattern test structure. Includes a first sub-resolution feature spaced from the feature.
Clause 13: The method of clause 12, wherein the one or more sub-resolution features are dimensioned to affect a corresponding pattern test structure differently than the first sub-resolution features. A second sub-resolution feature separated from the main feature by a second side different from the first side of the main feature by a second distance different from.
Clause 14: The method of any of clauses 11-13, wherein the one or more sub-resolution features are separated by a plurality of different discrete distances each dimensioned to affect a corresponding pattern test structure. Includes a plurality of sub-resolution features spaced from the main feature on the same side of the main feature.
Clause 15: The method of any of clauses 11-14, wherein the one or more sub-resolution features are separated from the side of the main feature by a first distance and from the opposite side of the main feature to the first side. A sub-resolution inversion feature disposed within the main feature to be separated by a second distance different from the distance.
Clause 16: The method of any of clauses 11-15, wherein the main feature comprises a substantially straight bar-shaped structure; the one or more sub-resolution features are on opposite sides of the main feature. Comprising a pair of small bar-shaped structures extending along the main feature at different distances from the feature, the width of the small bar being less than the resolution limit of the photolithographic patterning process, and the width of the main feature being the resolution. Greater than or equal to the limit; the reticle comprises a test grating including multiple instances of the test structure spaced apart from each other; the pattern test structure having different sloped sidewalls on opposite sides of the bar-shaped structure , The amount of difference between the tilts changes according to the change in focus or exposure amount.
(Item 17) The method according to any one of Items 11 to 16, wherein the substrate in the manufacturing process is patterned by photolithography with a reticle, and a pattern test is performed on the substrate in the manufacturing process along the pattern of at least a part of the manufacturing device. Manufacturing a structure; optically measuring at least a portion of the pattern test structure during the manufacturing process prior to completion of patterning of the manufacturing process; based on corresponding optical measurements and correlated models, Estimating the process characteristics of the photolithographic patterning of the manufacturing step; determining that the target process characteristics are different from the estimated process characteristics.
(Claim 18) A method of inferring a parameter of a patterning process, the method comprising: obtaining scattered radiation measurements of a patterned structure on a substrate; using one or more processors, a process of photolithographic patterning. Estimating a property based on an optical measurement using a calibrated model, the model comprising a first ratio relating a change in one of the process characteristics to a change in scattered radiation measurement. ..
(Item 19) The method according to item 18, wherein the first ratio is a partial derivative or reciprocal of the focus with respect to the scattered radiation measurement value, and the model is a partial derivative or reciprocal of the dose with respect to the scattered radiation measurement value. With a ratio of 2.
(Item 20) The method according to any one of items 18-19, comprising: acquiring training data; using the training data by determining the first ratio using one or more processors. Calibrating the model, comprising: calibrating the model by performing the model, wherein the training data is indicative of process characteristics of a photolithography process used to pattern a plurality of test structures on one or more substrates. A plurality of test conditions, in which test structures having different patterns are patterned under different test conditions, a pupil intensity measurement value by scatter measurement of the plurality of pattern test structures, and which measurement value is Data indicating whether or not the test condition is met.
Clause 21: The method of any of clauses 18-20, wherein the structure is a test structure, the method comprising: proximate the main feature and the main feature on a reticle, the method substantially comprising the main feature. Obtaining a reticle that at least partially defines the test structure having one or more sub-resolution features parallel to the reticle; photolithographically patterning one or more substrates with the reticle to form a patterned test structure on the substrate. Manufacturing.
22. The method of claim 21, wherein the one or more sub-resolution features are on the first side of the main feature a first distance dimensioned to affect a corresponding pattern test structure. A first sub-resolution feature spaced from the feature, the one or more sub-resolution features being dimensioned to affect a corresponding pattern test structure differently than the first sub-resolution feature. A second sub-resolution feature separated from the main feature by a second side different from the first side of the main feature by a second distance different from.
23. The method of claim 21, wherein the one or more sub-resolution features are separated from the side of the main feature by a first distance and are different from the opposite side of the main feature by the first distance. Includes sub-resolution inversion features located within the main feature so as to be separated by two distances.
(Item 24) The method according to any one of items 18-23, wherein the estimated process characteristic is different from the target value; and the photolithography patterning is performed so that the difference is reduced according to the determination. Adjusting process settings.
(Item 25) The method according to any one of items 18-24, which comprises manufacturing a plurality of electronic devices or optical devices on a substrate.
(Claim 26) A system comprising: one or more processors; and a memory that stores instructions that, when executed by at least a portion of the processors, cause the following steps to function: Obtaining training data comprising scattered radiation information, wherein the scattered radiation information is associated with each portion of the scattered radiation information at separate process conditions characteristic of the patterning process of each structure; using one or more processors And calibrating the model with the training data by determining a first ratio that relates a change in one of the process characteristics to a corresponding change in the scattered radiation information.
(Item 27) A tangible, machine-readable, non-transitory medium that stores instructions that, when executed by one or more processors, cause any of the steps of items 1-25 to function.
(Item 28) A system comprising: one or more processors; and a memory that stores instructions for executing the steps of any of items 1-25 when executed by at least some processors.

Claims (15)

A method of constructing a model for inferring process characteristics of the photolithography process from pupil intensity measurements by scatterometry of structures patterned in the photolithography process, comprising:
Acquiring training data including a plurality of pupil intensity measurements by scatterometry of each of the plurality of structures, and a plurality of process characteristics indicative of a process condition of a respective patterning process of the plurality of structures;
Determining using one or more processors a coefficient that correlates a change in at least one process characteristic with a corresponding change in the plurality of pupil intensity measurements. Determining a first coefficient that correlates a change in the first process characteristic with a corresponding change in the plurality of pupil intensity measurements;
Determining a second coefficient that associates a change in a second process characteristic with a corresponding change in the plurality of pupil intensity measurements,
The first coefficient is a partial differential with respect to the focus of the plurality of pupil intensity measurement values or a reciprocal of the partial differential ,
The method according to claim 1, wherein the second coefficient is a partial derivative with respect to a dose of the plurality of pupil intensity measurement values or an inverse number of the partial derivative . The method of claim 1, wherein the model is based on at least three Taylor series expansions. The model A method according to claim 1, characterized in that it comprises the inverse of the second-order partial differential or the second order partial derivative with respect to one of the process characteristics of the plurality of pupil intensity measurements. The method of claim 1, wherein building the model comprises determining model parameters using a Markov chain Monte Carlo algorithm. The method of claim 1, wherein building the model comprises determining model parameters using a Metropolis Hastings algorithm. Building the model comprises iteratively adjusting the model parameters to determine the model parameters based on an aggregate value of the agreement between each iteration of the model and at least a portion of the training data. The method of claim 1, comprising: Wherein the plurality of structures, using the reticle which at least partially defines a test structure, the focus - is patterned on one or more substrates according to the exposure amount matrix, different putter integrators strike structure different process conditions is patterned The method according to claim 1, wherein: The method of claim 1, wherein the training data comprises simulated training data obtained based on a model of a photolithography process and a model of scatterometry pupil intensity measurements. Obtaining a reticle at least partially defining a test structure having a main feature and one or more sub-resolution features proximate the main feature on the reticle and substantially parallel to the main feature;
Patterning one or more substrates with the reticle by photolithography. The one or more sub-resolution features include a first sub-resolution feature spaced from the main feature on a first side of the main feature by a first distance dimensioned to affect a corresponding pattern test structure;
The one or more sub-resolution features are dimensioned to affect the corresponding pattern test structure differently than the first sub-resolution features by a second distance different from the first distance. The method of claim 10 including a second sub-resolution feature spaced from the main feature on a second side different from the first side. The one or more sub-resolution features are spaced from the main feature on the same side of the main feature by a plurality of different distinct distances each dimensioned to affect a corresponding pattern test structure. 11. The method of claim 10, comprising: The one or more sub-resolution features are within the main feature to be spaced a first distance from one side of the main feature and a second distance different from the first distance from the opposite side of the main feature. 11. The method of claim 10 including sub-resolution inversion features located. The main feature comprises a substantially straight bar-shaped structure,
The one or more sub-resolution features comprise a pair of small bar-shaped structures extending along the main feature at different distances from the main feature on opposite sides of the main feature, the small bar width being Less than the resolution limit of the lithographic patterning process, the width of the main feature is greater than or equal to the resolution limit,
The reticle comprises a test grating including a plurality of the test structures spaced apart from each other,
11. The method of claim 10, wherein the pattern test structure has sidewalls of different slopes on both sides of the bar-shaped structure, and the amount of difference between the slopes changes in response to focus or exposure changes. .. One or more processors, and a memory that stores instructions that cause the following steps to function when executed by at least a portion of the processor, the steps comprising:
Acquiring training data including a plurality of pupil intensity measurements by scatterometry of each of the plurality of structures, and a plurality of process characteristics indicative of a process condition of a respective patterning process of the plurality of structures;
Constructing a model with the training data by determining a coefficient that correlates at least one process characteristic change with a corresponding change in the plurality of pupil intensity measurements using one or more processors; ,,
The system is characterized in that the model infers process characteristics of the photolithography process from pupil intensity measurements by scatterometry of structures patterned in the photolithography process .

Images